Pre-processing and EDA of Vaccinations data (India):

• The date of vaccinations available from 2020-01-15 till 2021-07-14
• Data Gathering of the below pre-processing and EDA steps from https://github.com/owid/covid-19-data/blob/master/public/data/vaccinations/country_data/India.csv
• Below are the initial set of insights from the above dataset
• This dataset is about the vaccination details specific to India
• Details about variables/columns:
  • location: Name of country for the data about vaccinations
  • date: Date of the observation
  • vaccine: List of vaccines administered in the country as per the current date shown
  • source_url: URL link for the data/record
  • total_vaccinations: Total number of doses administered. If a person receives one dose of the vaccine, this metric goes up by 1. If they receive a second dose, it goes up by 1 again (people_vaccinated + people_fully_vaccinated).
  • people_vaccinated: (one-dose vaccination) Total number of people who received at least one vaccine dose. If a person receives the first dose of a 2-dose vaccine, this metric goes up by 1. If they receive the second dose, the metric stays the same.
  • people_fully_vaccinated: (two-doses vaccination) Total number of people who received all doses prescribed by the vaccination protocol. If a person receives the first dose of a 2-dose vaccine, this metric stays the same. If they receive the second dose, the metric goes up by 1.
• This is a entire country-wise counts for the vaccination on a rolling daily basis of the above date range.
• Other factors of covid for India states-wise available @ https://www.kaggle.com/imdevskp/covid19-corona-virus-india-dataset?select=complete.csv but from 2020-01-30 till 2020-08-06

• Notebook by Ramya N (GitHub user name: rnedesigns)

import pandas as pd

import matplotlib
import matplotlib.pyplot as plt
%matplotlib inline

import plotly.express as px

To pre-process and explore datasets (EDA):

• Below steps to know about data fields, their content and data types
• Finding shape of the dataset, descriptive statistics, metadata about the DataFrame
• Imputing missing values if any
• Transform column based on the categorical columns or date field. Ex: Pandas provides date field as object type that requires to translate to datetime type especially for time-series analysis, encoding categorical values based on weighate of the variables for the model fitting
• Keeping the naming conventions of variables/fields consistant if they are not
• Segregating datasets to training and test sets

gh_raw_csv_url = 'https://raw.githubusercontent.com/owid/covid-19-data/master/public/data/vaccinations/country_data/India.csv'

vaccines_india_df = pd.read_csv(gh_raw_csv_url)
vaccines_india_df.head()

	location	date	vaccine	source_url	total_vaccinations	people_vaccinated	people_fully_vaccinated
0	India	2021-01-15	Covaxin, Oxford/AstraZeneca	https://twitter.com/MoHFW_INDIA/status/1350459...	0	0	0
1	India	2021-01-16	Covaxin, Oxford/AstraZeneca	https://twitter.com/MoHFW_INDIA/status/1350459...	191181	191181	0
2	India	2021-01-17	Covaxin, Oxford/AstraZeneca	https://twitter.com/MoHFW_INDIA/status/1350815...	224301	224301	0
3	India	2021-01-18	Covaxin, Oxford/AstraZeneca	https://www.mohfw.gov.in/	454049	454049	0
4	India	2021-01-19	Covaxin, Oxford/AstraZeneca	https://www.mohfw.gov.in/	674835	674835	0

# For number rows/records and fields
vaccines_india_df.shape

(178, 7)

vaccines_india_df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 178 entries, 0 to 177
Data columns (total 7 columns):
 #   Column                   Non-Null Count  Dtype 
---  ------                   --------------  ----- 
 0   location                 178 non-null    object
 1   date                     178 non-null    object
 2   vaccine                  178 non-null    object
 3   source_url               178 non-null    object
 4   total_vaccinations       178 non-null    int64 
 5   people_vaccinated        178 non-null    int64 
 6   people_fully_vaccinated  178 non-null    int64 
dtypes: int64(3), object(4)
memory usage: 9.9+ KB

# Here it's providing all the numerical computations in logarithmic format:
vaccines_india_df.describe()

	total_vaccinations	people_vaccinated	people_fully_vaccinated
count	1.780000e+02	1.780000e+02	1.780000e+02
mean	1.333785e+08	1.088657e+08	2.451280e+07
std	1.203281e+08	9.703832e+07	2.356104e+07
min	0.000000e+00	0.000000e+00	0.000000e+00
25%	1.504579e+07	1.241895e+07	2.626844e+06
50%	1.158584e+08	1.012642e+08	1.459415e+07
75%	2.150426e+08	1.712964e+08	4.374615e+07
max	4.049317e+08	3.218938e+08	8.303795e+07

round(vaccines_india_df.describe().T, 2)

	count	mean	std	min	25%	50%	75%	max
total_vaccinations	178.0	1.333785e+08	1.203281e+08	0.0	15045789.25	115858373.5	215042559.0	404931715.0
people_vaccinated	178.0	1.088657e+08	9.703832e+07	0.0	12418945.50	101264227.0	171296413.0	321893762.0
people_fully_vaccinated	178.0	2.451280e+07	2.356104e+07	0.0	2626843.75	14594146.5	43746146.0	83037953.0

# Convert Date column object type datetime type:

def datetime_split(df):
    df['date'] = pd.to_datetime(df['date'], format='%Y-%m-%d')
    df['year'] = df.date.dt.year
    df['month'] = df.date.dt.month

datetime_split(vaccines_india_df)

vaccines_india_df

	location	date	vaccine	source_url	total_vaccinations	people_vaccinated	people_fully_vaccinated	year	month
0	India	2021-01-15	Covaxin, Oxford/AstraZeneca	https://twitter.com/MoHFW_INDIA/status/1350459...	0	0	0	2021	1
1	India	2021-01-16	Covaxin, Oxford/AstraZeneca	https://twitter.com/MoHFW_INDIA/status/1350459...	191181	191181	0	2021	1
2	India	2021-01-17	Covaxin, Oxford/AstraZeneca	https://twitter.com/MoHFW_INDIA/status/1350815...	224301	224301	0	2021	1
3	India	2021-01-18	Covaxin, Oxford/AstraZeneca	https://www.mohfw.gov.in/	454049	454049	0	2021	1
4	India	2021-01-19	Covaxin, Oxford/AstraZeneca	https://www.mohfw.gov.in/	674835	674835	0	2021	1
...	...	...	...	...	...	...	...	...	...
173	India	2021-07-13	Covaxin, Oxford/AstraZeneca, Sputnik V	https://www.mohfw.gov.in/	387697935	311387539	76310396	2021	7
174	India	2021-07-14	Covaxin, Oxford/AstraZeneca, Sputnik V	https://www.mohfw.gov.in/	391340491	313529502	77810989	2021	7
175	India	2021-07-15	Covaxin, Oxford/AstraZeneca, Sputnik V	https://www.mohfw.gov.in/	395343767	316116189	79227578	2021	7
176	India	2021-07-16	Covaxin, Oxford/AstraZeneca, Sputnik V	https://www.mohfw.gov.in/	399695879	318665226	81030653	2021	7
177	India	2021-07-17	Covaxin, Oxford/AstraZeneca, Sputnik V	https://www.mohfw.gov.in/	404931715	321893762	83037953	2021	7

178 rows × 9 columns

# Check date has been converted to datetime object:
vaccines_india_df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 178 entries, 0 to 177
Data columns (total 9 columns):
 #   Column                   Non-Null Count  Dtype         
---  ------                   --------------  -----         
 0   location                 178 non-null    object        
 1   date                     178 non-null    datetime64[ns]
 2   vaccine                  178 non-null    object        
 3   source_url               178 non-null    object        
 4   total_vaccinations       178 non-null    int64         
 5   people_vaccinated        178 non-null    int64         
 6   people_fully_vaccinated  178 non-null    int64         
 7   year                     178 non-null    int64         
 8   month                    178 non-null    int64         
dtypes: datetime64[ns](1), int64(5), object(3)
memory usage: 12.6+ KB

vaccines_india_df.groupby('month').agg({'total_vaccinations': 'median', 
                                        'people_vaccinated': 'median', 
                                        'people_fully_vaccinated': 'median'})

	total_vaccinations	people_vaccinated	people_fully_vaccinated
month
1	1582201.0	1582201.0	0.0
2	8516771.0	8418653.0	98118.0
3	35064536.0	28862037.0	6202499.0
4	114493238.0	100179199.0	14314039.0
5	181897771.0	141448223.5	40449547.5
6	256815633.0	209136960.5	47678672.5
7	372511904.0	300925205.5	71586698.5

# So the dataset from resource contains mean value for total_vaccinations, people_vaccinated & people_fully_vaccinated for every day:
vaccines_india_df.groupby('month').agg({'total_vaccinations': 'mean', 
                                       'people_vaccinated': 'mean', 
                                       'people_fully_vaccinated': 'mean'})

	total_vaccinations	people_vaccinated	people_fully_vaccinated
month
1	1.666071e+06	1.666071e+06	0.000000e+00
2	8.806005e+06	8.190648e+06	6.153570e+05
3	3.761312e+07	3.147027e+07	6.142853e+06
4	1.125233e+08	9.670344e+07	1.581990e+07
5	1.810848e+08	1.432658e+08	3.781903e+07
6	2.643352e+08	2.154857e+08	4.884952e+07
7	3.692846e+08	2.987195e+08	7.056514e+07

# The y-axis columns are of mean values from this dataset:

plt.bar(vaccines_india_df.month, vaccines_india_df.total_vaccinations, label='Total Vaccines Administered')
plt.bar(vaccines_india_df.month, vaccines_india_df.people_vaccinated, label='Once Vaccinated')
plt.bar(vaccines_india_df.month, vaccines_india_df.people_fully_vaccinated, label='Fully Vaccinated')

plt.title('Total Vaccines VS Once Vaccinated VS Fully Vaccinated', fontsize=14, y=1.08)
plt.xlabel('Monthly')
plt.ylabel('Number of People')
plt.legend();

# df = pd.read_csv('../task-1_datasets/collected_datasets/india_vacc_data-01152021-16072021.csv', parse_dates=['date'])
# df.info()
# df.head()

px.bar(vaccines_india_df, 
       x=vaccines_india_df.month, 
       y=vaccines_india_df.total_vaccinations, 
       title="Total Vaccines Administered from 2021-01-15", 
       color="vaccine", 
       hover_data=['date'])

px.bar(vaccines_india_df, 
       x=vaccines_india_df.month, 
       y=vaccines_india_df.people_vaccinated, 
       title="Total People Vaccinated Once from 2021-01-15", 
       color="vaccine", 
       hover_data=['date'])

fig = px.bar(vaccines_india_df, 
             x=vaccines_india_df.month, 
             y=vaccines_india_df.people_fully_vaccinated, 
             title="Total People Fully Vaccinated from 2021-01-15", 
             color="vaccine", 
             hover_data=['date'])
# labels=dict(x='Monthly - 2021', y='Pople Fully Vaccinated', color='Vaccines Legend')
fig.show()

"""
For comparing all three Total Vaccines VS Once Vaccinated VS Fully Vaccinated
in one Graph:
"""

import plotly.graph_objects as go

# Stacked Bar Graph

fig = go.Figure(data=[
    go.Bar(name='Total Vaccines Administered', 
           x=vaccines_india_df.month, 
           y=vaccines_india_df.total_vaccinations, 
           hovertext=vaccines_india_df.date), 
    go.Bar(name='People Once Vaccinated', 
           x=vaccines_india_df.month, 
           y=vaccines_india_df.people_vaccinated, 
           hovertext=vaccines_india_df.date), 
    go.Bar(name='People Fully Vaccinated(Two doses)', 
           x=vaccines_india_df.month, 
           y=vaccines_india_df.people_fully_vaccinated, 
           hovertext=vaccines_india_df.date)
])

fig.update_xaxes(title='Monthly From January 2021')
fig.update_yaxes(title='Total Vaccines Administered, Once Vaccinated & Fully Vaccinated', 
                 title_font={'size': 12})
fig.update_layout(barmode='stack', 
                  title='Total Vaccines Administered VS Once Vaccinated VS Fully Vaccinated')
fig.show()